Share

Performance prediction and calibration of a clean-room air conditioner using an embedded artificial neural network.

Number: 2106

Author(s) : YOON M. S., YI D. H., SEO M.K., RYU S. Y.

Summary

This study is about the application of an artificial neural network (ANN) to implement supervised learning for the performance prediction of a clean-room air conditioner (CRAC) installed on-site. To measure accurately the cooling capacity and efficiency of an HVAC product such as an air conditioner, the temperature and humidity should be fixed in a well-defined standard chamber. However, at an actual site where air conditioners are installed, it is unreasonable to expect a well-defined testing chamber environment. To resolve this difficulty, various temperature and humidity environments were simulated under the laboratory conditions in advance. Moreover, the sensing and performance data measured by the sensors inside of the CRAC product were recorded along with the data measured in an air enthalpy-type standard chamber. After simultaneous acquisition of the CRAC and standard-chamber data in a simulated chamber environment, supervised learning by an artificial neural network was carried out and the trained ANN was transferred into an embedded chipset. Finally, accuracy analyses of the control-group ANN (using chamber sensors) and experimental-group ANN (using product sensors) are compared for selected test conditions. Although the experimental-group ANN shows worse prediction performance than the control-group ANN does, it shows better results than the product calculation results. The experimental-group ANN of the CRAC might exhibit prediction as good as the control-group ANN, if the precision of the product sensors is improved.

Available documents

Format PDF

Pages: 10 p.

Available

Free

Download

Details

  • Original title: Performance prediction and calibration of a clean-room air conditioner using an embedded artificial neural network.
  • Record ID : 30030467
  • Languages: English
  • Subject: Technology
  • Source: 2022 Purdue Conferences. 19th International Refrigeration and Air-Conditioning Conference at Purdue.
  • Publication date: 2022/07/10

Links


See other articles from the proceedings (224)
See the conference proceedings

Indexing